Talks: Open Source Industrial Control: Turning 2,800 Tons of Metal with Python and Flask

Saturday - May 18th, 2024 4:15 p.m.-4:45 p.m. in Hall C

Presented by:

Description

This is a story of how Python can fit into the physical world around us. It is a story of system design and product development. It is a tale of great breadth, covering distributed computing, custom printed circuits, electromagnetism, some of the largest hydropower generators in the world, and the software and hardware that brings this all together. This tale covers several years of research and development, culminating in a cyber physical system built on open-source software and easily attainable off-the-shelf products and components.

The Bureau of Reclamation is the 2nd largest hydropower generator in the United States and operates 53 hydroelectric plants with 14.7 gigawatts of installed capacity. On average, Reclamation plants generate 40 billion kilowatt hours of electricity annually. It takes a large fleet of rotating machines (generators and pumps) across the western United States to produce this much electricity, and operations and maintenance of this fleet is a significant part of Reclamation’s annual budget.

During routine diagnostic and maintenance tasks, it is often necessary to slowly and precisely turn a machine’s rotor to make measurements at various rotational positions. This was previously done with many personnel climbing into the machine and pushing (or, for very large machines, complicated crane maneuvers). This is dangerous, inefficient, and lacks precision.

With custom software and intelligent system design, we have developed a modular system that can easily scale to turn even our largest machines with the click of a button. The software can track the machine’s rotational position in real time to one thousandth of a degree and can easily expand to accommodate future diagnostic test methods. What once put personnel at risk and potentially took days to carry out can now be done in hours with little to no risk to personnel.